updates

app.py

@@ -75,11 +75,19 @@ def translate(video_file):
     return {"translation":translation}
 
 # Gradio App config
-title = "ASL Translation (MoViNet + T5)"
-description =   "Gradio demo of word-level sign language classification using I3D model pretrained on the WLASL video dataset. " \
-                "WLASL is a large-scale dataset containing more than 2000 words in American Sign Language. " \
-                "Examples used in the demo are videos from the the test subset. "  \
-                "Note that WLASL100 contains 100 words while WLASL2000 contains 2000."
+title = "American Sign Language Translation: An Approach Combining MoViNets and T5"
+
+description =   """
+This application surfaces a model for translation of American Sign Language (ASL), 
+which comprises of a fine-tuned MoViNets CNN model and a T5 encoder-decoder model 
+to generate translations from the video embeddings. This model architecture achieves a BLEU score of 1.98 
+and an average cosine similarity score of 0.21 when trained and evaluated on the YouTube-ASL dataset. 
+More information about the models can be found in our GitHub repository <a href=https://github.com/deanna-emery/ASL-Translator>here</a>.
+
+A limitation of this architecture is the size of the MoViNets model, making it especially slow during inference on a CPU. 
+We do not recommend uploading videos longer than 4 seconds as the video embedding generation may take some time. 
+We have provided some pre-cached videos with their original captions and translations as examples.
+"""
 
 
 examples = [
@@ -92,9 +100,16 @@ examples = [
 
 article =  "More information about the models can be found <a href=https://github.com/deanna-emery/ASL-Translator>here</a>."
 
+article =  """The captions for the example videos are as follows in order: \n
+1. 'My second ASL professor's name was Will White'
+2. 'You are my sunshine'
+3. 'scrub your hands for at least 20 seconds'
+4. 'no'
+"""
+
 # Gradio App interface
 gr.Interface(fn=translate,
-              inputs="video", 
+              inputs=["video", 'text'], 
               outputs="text",
               allow_flagging="never",
               title=title,